In the past year, we largely worked on four projects. Because rats learn to lever-press for brief electrical stimulation of the median and dorsal raphe nuclei (MRN and DRN, respectively), these brain sites have long been implicated in reward processes. However, it is not clear whether the MRN and DRN integrate reward-related signals or merely contain fibers of passage involved in reward processes. To shed light on this issue, the present study employed chemicals that selectively modulate neurotransmission, in particular the GABAA receptor agonist muscimol. Rats quickly learned to lever-press for muscimol infusions (50 and 100 M) into the MRN or DRN. Muscimol was not self-administered when cannulae were placed just outside these nuclei. The reinforcing effects of muscimol appeared to be greater when the drug was administered into the MRN than into the DRN, as demonstrated by higher infusion rates and better response discrimination. These observations are consistent with the additional finding that muscimol administration into the MRN, but not the DRN, induced conditioned place preference. The reinforcing effects of muscimol administration into the MRN were blocked by co-administration of the GABAA antagonist picrotoxin (100 M) and by pretreatment with the dopamine receptor antagonist SCH 23390 (0.025 mg/kg, i.p.). The present results suggest that median and dorsal raphe neurons via GABAA receptors are involved in integration of primary reinforcement, and that median raphe neurons, presumably inhibited by muscimol administration, may exert tonic inhibition over dopamine-dependent reward circuitry. The midbrain raphe nuclei may be involved in a variety of reward-related phenomena including drug addiction. [unreadable] [unreadable] Our recent data demonstrated that selective regions of the ventral striatum, such as the medial accumbens shell and medial olfactory tubercle, but not accumbens core, lateral accumbens shell, or lateral olfactory tubercle supported self-administration of cocaine and amphetamine delivered via intracranial infusions. These findings suggest that the ventromedial striatum is more important than the ventrolateral striatum in mediating rewarding effects of certain drugs of abuse. The present experiment sought to investigate whether intracranial self-administration of ()-3,4-methylenedioxymethamphetamine (MDMA) into these same forebrain regions would elicit similar rewarding behaviors. Over four consecutive days, rats were trained to lever press for 30mM MDMA or vehicle infusions for a period of 90 min each day. Subsequently, on three separate days, an increasing dosage of MDMA (10, 30, 100mM) or vehicle was presented for self-administration. Rats learned to self-administer MDMA at very moderate rates into medial accumbens shell, medial olfactory tubercle and accumbens core, but did not self-administer it into the lateral shell or lateral tubercle. In a separate experiment, very high dosages of MDMA (500mM, 1M) over several sessions followed by amphetamine were examined. Rats learned to self-administer MDMA into the medial shell at similar rates as they did d-amphetamine into the same site. On the other hand, rats did not learn to self-administer MDMA into the medial tubercle or core, even though medial tubercle rats did learn to self-administer d-amphetamine into the same site. Taken together, these data suggest the rewarding effects of MDMA is most effectively mediated by the medial shell among ventral striatal regions. [unreadable] [unreadable] Neuropeptide S (NPS) is a recently identified neuropeptide. NPS-containing cell bodies are localized in the lower brainstem just adjacent to the locus coeruleus and its receptors are expressed in the brain areas that are implicated in motivation and reward. Indeed, NPS appears to play an important role in arousal. We examined possible rewarding effects of NPS injected into the lateral ventricle, using intracranial self-administration and conditioned place preference procedures. Each male Wistar rat was implanted with a permanent unilateral guide cannula ended just above the lateral ventricle. For intra-ventricular self-administration, rats received NPS at the dose of 3.39 nmol per infusion in sessions 2-4 and vehicle in sessions 1 and 5 followed by three sessions with 0.339, 3.39 and 33.9 nmol doses. Each session lasted for 90 min, and sessions were separated by 24 hours. Rats learned quickly to self-administer the 3.39 nmol dose into the lateral ventricle. It is unclear whether rats self-administer NPS at a lower dose. Effects of two doses of NPS were evaluated for conditioned place preference. Each rat was confined in a compartment for 20 min immediately after intraventricular injections of NPS (100 or 1,000 nmol) and confined in another compartment for 20min after vehicle injections. Injections were separated by 24 hours. Pairings of NPS and vehicle injections with these compartments were alternated over 8 days. Place preference of each rat was examined in the first session, prior to conditioning, and the last session, after conditioning, by allowing them to have free access to both compartments for 15 min without injections. The high dose (1,000 nmol) of NPS induced conditioned place preference, whereas the low dose (100 nmol) induced conditioned place avoidance. These data suggest that NPS injections into the lateral ventricle are rewarding, although NPS injections at some doses may also be aversive. These findings and the previous finding that intra-ventricular injections of NPS elicit motor stimulant effects raise the possibility that the NPS system interacts with the mesolimbic dopamine system. [unreadable] [unreadable] Finally, we reviewed anatomical and functional refinements of the meso-limbic dopamine system of the rat in the paper published in Brain Research Reviews. Present experiments suggest that dopaminergic neurons localized in the posteromedial ventral tegmental area (VTA) and central linear nucleus raphe selectively project to the ventromedial striatum (medial olfactory tubercle and medial nucleus accumbens shell), whereas the anteromedial VTA has few if any projections to the ventral striatum, and the lateral VTA largely projects to the ventrolateral striatum (accumbens core, lateral shell and lateral tubercle). These findings complement the recent behavioral findings that cocaine and amphetamine are more rewarding when administered into the ventromedial striatum than into the ventrolateral striatum. Drugs such as nicotine and opiates are more rewarding when administered into the posterior VTA or the central linear nucleus than into the anterior VTA. A review of the literature suggests that: (1) the midbrain has corresponding zones for the accumbens core and medial shell; (2) the striatal portion of the olfactory tubercle is a ventral extension of the nucleus accumbens shell; (3) a model of two dopamine projection systems from the ventral midbrain to the ventral striatum is useful for understanding reward function. The medial projection system is important in the regulation of arousal characterized by affect and drive, and plays a different role in goal-directed learning than the lateral projection system, as described in the variation-selection hypothesis of striatal functional organization.